AI EngineeringApr 3, 2026
Gemma 4 and Cursor 3 Drop Today: A Dev Roundup
Google's most capable open models hit the streets while Cursor goes all-in on agents. Here's what matters for developers.
TobWriting
Thoughts on backend architecture, AI systems, tech, and life.
TobAI EngineeringApr 2, 2026
Axios Got Backdoored: What Developers Need to Know About the npm Supply Chain Attack
The popular HTTP client library Axios was compromised with a malicious dependency, affecting versions 1.14.1 and 0.30.4. Here's what happened and how to protect yourself.
TobAI EngineeringApr 1, 2026
The Day the Code Broke: Claude Code Leaks and the axios Supply Chain Attack
Two developer bombshells dropped this week: Anthropic accidentally shipped Claude Code's source map to npm, and attackers compromised axios with a precision supply chain strike. Here's what actually happened and what it means for you.
TobTutorialsApr 1, 2026
How to Start Development with Agentic Workflow
A practical guide for developers stepping into agentic workflow development. Learn the core concepts, tooling, and step-by-step approach to building your first agentic system — from local setup to production-ready architecture.
TobAI EngineeringMar 30, 2026
AI Tooling This Week: Supply Chain Attacks Get Smarter and Enterprise Agents Go Private
A malicious PyPI package, Cursor's self-hosted coding agents, and why Matt Webb thinks vibe coding is making us better architects.
TobAI EngineeringMar 30, 2026
AI This Week: A Victorian Chatbot, a PyPI Supply Chain Attack, and Text Rendering Magic
A LLM trained on 19th-century British novels, a malware-infected Python package, and a clever new library for measuring text in browsers. Here's what's been cooking in AI this week.
TobAI EngineeringMar 28, 2026
AI Sycophancy and the Rise of Vibe Coding: A Reality Check
Two big themes are colliding in AI development right now: models that tell you what you want to hear, and developers who code by vibes instead of line counts. Here's what's actually going on.
TobAI EngineeringMar 27, 2026
Someone Found a Live Malware Campaign on PyPI by Accident
A developer debugging a frozen laptop stumbled onto litellm 1.82.8, a weaponized package designed to steal SSH keys, AWS credentials, and just about everything else on your machine.
TobAI EngineeringMar 27, 2026
Can MiniMax M2.7 Replace Claude Sonnet or Opus? A Developer's Honest Take
MiniMax M2.7 now supports the Anthropic SDK with a two-line config swap. But can it actually replace Claude Sonnet or Opus in your production stack? Here's what the benchmarks and real-world gaps tell you.
TobAI EngineeringMar 27, 2026
The Day AI Found Its Own Malware: LiteLLM Attack and the Cognitive Debt Crisis
A developer using Claude Code discovered they were patient zero for a PyPI supply chain attack. Meanwhile, the creator of a major AI agent framework is warning that we are building unmaintainable codebases at record speed. Two stories, one uncomfortable truth.
TobAI EngineeringMar 26, 2026
LiteLLM Supply Chain Attack and Claude Code Auto Mode
The LiteLLM PyPI incident that hit 47k downloads, plus Claude Code's new permissions safety net.
TobDevOps & ToolsMar 25, 2026
Series: What is OpenClaw?
OpenClaw is an AI agent platform that goes beyond simple chatbots. This post covers its architecture, the different agent types, how the workspace works, and why it is built for real development workflows.
TobToolsMar 25, 2026
OpenCode.ai CLI Cheatsheet
Quick reference for OpenCode CLI commands and flags.
TobAI EngineeringMar 25, 2026
How to Create Agent Skills: CLI API Tester Example
Skills let you package domain expertise into reusable instruction packs that agents can auto-load. Here's how to build one from scratch, using a CLI API Tester skill as a real example.
TobAI EngineeringMar 25, 2026
AI Roundup: Litellm Supply Chain Hack, Massive MoE Models on Your Laptop, Cursor Gets Serious
A compromised PyPI package steals SSH keys, researchers run trillion-parameter models on 96GB RAM, and Cursor ships Composer 2 with automations.
TobAI EngineeringMar 24, 2026
iPhone 17 Pro Runs 400B LLM, Walmart's ChatGPT Flops, and Starlette 1.0 Drops
Plus the new Python web framework that's quietly powering FastAPI, and what Walmart's failed experiment tells us about real-world AI adoption.
TobAI EngineeringMar 24, 2026
The AI Dev Digest: LiteLLM Supply Chain Hack, llama.cpp Joins Hugging Face
This week's AI news that matters: a critical PyPI supply chain attack hits LiteLLM, llama.cpp officially joins Hugging Face, and massive MoE models start running on your MacBook.
TobAI EngineeringMar 23, 2026
AI Roundup: Cursor's Kimi-Powered Composer, OpenAI's Astral Acquisition, and Running 397B on a Laptop
Three big stories from the AI world: Cursor's new Composer 2 uses Kimi K2.5, OpenAI just bought the company behind uv and ruff, and someone got a 397B model running on a laptop.
TobAI EngineeringMar 21, 2026
AI Roundup: OpenAI Buys Astral, Qwen 397B on MacBook, Cursor Gets Smarter
Big moves in the AI world this week: OpenAI acquires the Python tooling powerhouse, someone gets a 397B model running on a laptop, and Cursor drops Composer 2 powered by Kimi.
TobAI EngineeringMar 20, 2026
OpenAI Just Bought the Company Behind uv and Ruff
The biggest Python tooling company is now part of OpenAI. What this means for the open source ecosystem and the future of coding.
TobAI EngineeringMar 20, 2026
Agentic AI: The Complete Guide to Autonomous AI Systems in 2026
Agentic AI is reshaping how software is built. Learn what agentic AI systems are, how they work, the core architectural patterns, real-world use cases, and what engineers need to know to build reliable autonomous agents in production.
TobAI EngineeringMar 18, 2026
AI Roundup: Snowflake Sandbox Break, Python 3.15 JIT Beats Expectations, and Cursor's Plugin Power Play
Security flaw in Snowflake's AI agent, Python's JIT ahead of schedule, and Cursor's massive plugin ecosystem expansion.
TobAI EngineeringMar 18, 2026
AI Roundup: Mistral's New MoE Model and the Rise of Subagent Patterns
Mistral drops a massive 119B parameter model while the community figures out how to scale AI coding agents without blowing through context limits.
TobAIMar 17, 2026
Why Small Language Models Are the Future of On-Device AI
Big models get the buzz, but tiny models are where the real revolution is happening.
TobToolsMar 17, 2026
Claude Code vs Cursor: Which AI Coding Assistant Actually Works?
A practical comparison of two leading AI coding tools—Anthropic's CLI agent and Cursor's IDE integration.
TobDevOpsMar 17, 2026
AI Coding Assistant Best Practices: Getting the Most from Your AI Pair Programmer
Prompt engineering is just the start. Here's how to actually integrate AI assistants into your development workflow effectively.
TobAIMar 17, 2026
AI Agent Design Patterns Every Developer Should Know
From reflex agents to planning agents, here's a comprehensive breakdown of the architecture patterns powering modern AI systems.
TobAI EngineeringMar 16, 2026
Claude's 1M Context Drop + NVIDIA's Vera CPU: The Agentic AI Week
Anthropic just made long-context AI affordable, while NVIDIA drops a surprise CPU built specifically for AI agents. Here's what matters for developers.
TobAI EngineeringMar 15, 2026
AI Roundup: Claude Hits 1M Context While Shopify Uses AI Agents to Speed Up Liquid by 53%
Anthropic makes 1M context generally available at no premium pricing, and Shopify's CEO Tobi Lütke uses AI coding agents to dramatically optimize their Liquid template engine.
TobAI EngineeringMar 13, 2026
Claude Hits 1M Context While Shopify CEO Uses AI Agents to 2x Performance
Anthropic drops the long-context premium, Tobi Lütke leverages AI coding agents for a 53% Liquid speed boost, and Cursor expands its plugin ecosystem.
TobAIMar 13, 2026
How AI Coding Agents Helped Shopify Achieve 53% Performance Boost
Tobias Lütke used AI coding agents to squeeze 53% faster performance out of Shopify's Liquid template engine. Here's what the autoresearch pattern reveals about the future of programming.
TobAI EngineeringMar 13, 2026
AI Coding Agents Just Made Shopify's Template Engine 53% Faster
Tobias Lutke used AI agents to optimize Liquid, Shopify's 20-year-old template engine. The results are eye-opening for anyone wondering what AI coding agents can actually accomplish.
TobAI EngineeringMar 8, 2026
My Saturday Is Gone. Yours Should Be Too. Here's Why.
Hujan dari pagi, nggak bisa keluar, ngk jadi beli baju lebaran. Akhirnya duduk, buka laptop, dan mulai explore hal-hal baru yang udah lama ada di list. Ini ceritanya.
TobAI EngineeringMar 8, 2026
AI Engineering Updates: GPT-5.4, Claude Max for OSS, and Agentic Testing
OpenAI drops GPT-5.4 with a massive context window, Anthropic gives Claude Max to open source maintainers, and why agentic testing is the new standard.
TobAI EngineeringMar 7, 2026
Clinejection: How One GitHub Issue Title Could Have Pushed Malware to Millions of Developers
A researcher found a prompt injection vulnerability in Cline's AI-powered issue triage that could have compromised production releases on VS Code Marketplace and OpenVSX. The attack chain is a masterclass in how GitHub Actions trust boundaries break down.
TobSeries - Build Your Own Background AgentMar 6, 2026
Series: Apa Itu Background Agent dan Kenapa Harus Bangun Sendiri?
Ramp membangun coding agent internal bernama Inspect — dan 30% dari semua PR mereka sekarang ditulis oleh agent ini. Bukan pakai Copilot, bukan pakai Cursor. Mereka bangun sendiri. Kenapa?
TobAI EngineeringMar 6, 2026
Cursor Just Became an Always-On Agent Platform
Cursor released two major updates this week: Automations for always-on agents triggered by Slack, GitHub, PagerDuty, and more — plus Cursor is now available inside JetBrains IDEs via ACP.
TobAI EngineeringMar 6, 2026
AI Roundup: GPT-5.4 Is Out, GitHub Issue Attack Hit 4K Devs, and AI Just Rewrote Open Source Law
OpenAI dropped GPT-5.4. A malicious GitHub issue title compromised 4,000 developer machines. And a coding agent rewrote a Python library to change its license — now the original author is fighting back.
TobAIMar 5, 2026
Today in AI: Qwen Momentum, NanoGPT Slowrun, and Companion Risks
A quick look at today's top AI discussions, focusing on Qwen's latest moves, NanoGPT experiments, and the real-world impact of AI companions.
TobAI EngineeringMar 5, 2026
AI Roundup: Gemini 3.1 Flash-Lite, Donald Knuth Praises Claude, and Qwen Drama Continues
Google drops a budget model that beats its older flagship. Donald Knuth just had an open math problem solved by Claude Opus 4.6. And the Qwen team fallout keeps getting messier.
TobAI EngineeringMar 4, 2026
Qwen's Lead Researcher Just Quit. What It Means for Open Source AI
Junyang Lin, the technical lead behind Alibaba's Qwen models, announced his resignation today. Several core team members followed. Here's what happened and why it matters for the open source AI space.
TobAI EngineeringMar 4, 2026
Does AI Make You a Worse Developer? Anthropic's Study Has Answers
Anthropic ran a randomized controlled trial on 52 software engineers to find out if AI coding tools hurt skill development. The results are worth reading before you hand everything to Copilot.
TobAI EngineeringMar 4, 2026
AI Roundup: GPT-5.3 Instant, Knuth's Paper on Claude, and the Chatbot Backlash
Three things the dev community is talking about today: OpenAI drops GPT-5.3 Instant, Don Knuth writes about Claude at Stanford, and developers are pushing back on forced chatbot UIs.
TobAI EngineeringMar 1, 2026
AI Agent Design Patterns: Panduan Lengkap Memilih Arsitektur yang Tepat
Membangun sistem AI agentic bukan cuma soal model yang canggih — arsitektur yang tepat menentukan segalanya. Panduan lengkap 12 pola desain agen dari Google Cloud untuk membangun sistem AI yang andal dan efektif.
TobAI EngineeringMar 1, 2026
AI Agent untuk Backend Developer (Versi Bahasa Bayi 👶)
Pusing baca dokumentasi AI Agent yang ribet? Sini, kita bahas pakai bahasa bayi. Anggap aja AI Agent itu robot mainan pintar yang bisa disuruh beresin balok.
TobSeries - Coding AgentsFeb 27, 2026
Series: Mengenal Coding Agents: Era Baru Software Engineering
Stripe merge 1.300 pull request per minggu yang 100% ditulis AI. Tidak ada satu baris kode manusia. Ini bukan masa depan, ini sudah terjadi sekarang, dan kita perlu memahaminya.
TobSeries - Became an AI EngineerFeb 25, 2026
Series: What is an AI Engineer?
AI Engineer is not a data scientist, not an ML researcher. It's a new role that sits between software engineering and AI — and it's one of the most in-demand jobs right now.
TobSeries - API Gateway & MicroservicesFeb 25, 2026
Series: What is an API Gateway?
In a microservices world, every service speaks for itself — but someone has to manage the front door. That someone is the API Gateway. Let's break down what it is, why it exists, and how it fits into modern distributed systems.
TobSoftware EngineeringFeb 25, 2026
POSSE: Strategi Publish Konten yang Bikin Kamu Punya Kendali Penuh
Pernah nulis artikel di Medium, tapi Medium tutup besok? Kenalan sama POSSE - strategi publish konten yang bikin kamu tidak bergantung pada platform manapun.
TobSoftware EngineeringFeb 25, 2026
KISS, YAGNI, DRY: Tiga Prinsip yang Wajib Diketahui Setiap Developer
Tiga prinsip sederhana yang bisa membuat kode kita lebih bersih, mudah dibaca, dan tidak over-engineering. Cocok untuk semua level developer.
TobAIFeb 25, 2026
Google Just Solved The Greatest Limitation of AI Agents
AI agents have always struggled with one fundamental problem: memory. Here's how Google is changing that.
TobAI EngineeringFeb 25, 2026
Building AI Agents: Patterns and Anti-Patterns (A Production Guide)
Most AI agents fail in production because of architecture mistakes, not LLM limitations. Here are the 5 patterns that work and the 5 anti-patterns that will break your system.
TobSeries - Backend MasteryFeb 24, 2026
Series: What is Backend Development?
Every app you love has an invisible engine powering it. Let's pull back the curtain on backend development — what it is, what it does, and why it matters.
TobDocumentationFeb 21, 2026
Visualizing Workflow Architectures with Mermaid
A practical guide to documenting complex backend systems and state machines using Mermaid diagrams directly in your markdown files.
TobArchitectureJan 14, 2026
Designing Resilient API Layers for AI-Native Applications
A deep dive into building fault-tolerant API architectures that gracefully handle the unpredictable nature of large language model integrations.
TobPerformanceJan 10, 2026
Semantic Caching: Beyond Simple Key-Value Lookups
Traditional caching fails with AI because slight variations in prompts produce identical intents. Here's how to implement semantic similarity caching.
TobPatternsJan 5, 2026
Event Sourcing for AI Workflow Orchestration
How event sourcing patterns enable reproducible, debuggable AI pipelines with full auditability and time-travel debugging capabilities.
TobInfrastructureDec 28, 2025
Vector Databases in Production: Lessons Learned
Real-world insights from deploying vector search at scale, including indexing strategies, query optimization, and infrastructure considerations.
TobFrontendDec 15, 2025
Streaming LLM Responses: From Tokens to UI
End-to-end patterns for implementing real-time streaming from language models to user interfaces with proper error handling.
TobSeries - Build MCP ServerJan 12, 2025
Series: Apa itu MCP Server? Panduan Lengkap Model Context Protocol untuk Developer
Pelajari apa itu MCP Server (Model Context Protocol), kenapa penting untuk AI, dan cara kerjanya. Mulai bangun MCP server pertamamu di series ini.
Tob